Interrogation system employing prior knowledge about an object to discern an identity thereof

This application is a continuation of patent application Ser. No. 11/238,225, entitled “Interrogation System Employing Prior Knowledge About An Object To Discern An Identity Thereof,” filed on Sep. 29, 2005, which application is incorporated herein by reference.

The present invention is directed, in general, to communication systems and, more specifically, to an interrogation system employing prior knowledge about an object to discern an identity thereof and methods of operating the same.

Present state of the art transponders, though performing a valuable operation, still often suffer from a condition of not having the responses reliably detected. The reasons for lack of reliable detectable responses are due to a number of factors such as attenuation and corruption of the signal due to various materials absorbing, reflecting, or scattering the signal, as well as noise and interference due to other radio frequency energy sources, noise and interference due to a plurality of radio frequency signals, and other problems familiar to those skilled in the art. As a result, a number of process steps, which might otherwise be automated, still employ manual and costly additional process steps to fully obtain the desirable information. An example is the process of receiving goods at a warehouse. A quantity of goods, each with radio frequency identification (“RFID”) transponders or tags affixed thereto is stacked on a pallet. Additionally, an RFID tag is also affixed to the pallet. The pallet is then moved through a doorway surrounded by antennas. The antennas are the apertures for sending an interrogation signal to the RFID tags, as well as the apertures for receiving a reply. The radio frequency characteristics of the goods, such as the dielectric characteristics of the product and packaging, may often obstruct the signal path such that items within the interior of the stack are not detected. Thus, an accurate inventory count requires manual removal of all goods from the pallet requiring additional personnel and time. This amounts to hundreds of thousands of dollars per year in lost productivity for each such doorway or portal.

Many fields of commerce involve the need for counting, tracking, or accounting for items. In general, two of the lowest cost approaches presently in use involve bar code scanning or RFID techniques. Bar code scanning in general requires a clear optical path between an interrogator and the bar code while RFID techniques, which is a varied collection of technical approaches for many applications, in general does not. As with bar code scanning, the RFID techniques are primarily used for automatic data capture, and have the potential to significantly alter how processes occur and how companies operate. Bar code scanning alone, primarily because of the requirement of a clear optical path as well as limited information transfer capacity, is generally not compatible with many automated processes, such as the example of receiving goods previously cited.

One familiar application of RFID techniques, highly visible to travelers and consumers, is “smart labels” seen in airline baggage tracking, and in many stores for inventory control and for deterrence of theft. In some cases, the smart labels may combine both RFID and bar coding techniques. The RFID tags are in general very versatile and can be adapted for many and diverse applications. The RFID tags can be configured with or without batteries; and can be read only or read/write. The RFID tags can also be configured to operate over various ranges of separation between the interrogator and the tag. Less familiar, but very common applications include the inclusion of RFID tags into automobile key fobs as anti-theft devices, into identification badges for employees, and RFID tags incorporated into wrist bands as an accurate and secure method of identifying and tracking hospital patients, prison inmates, and patrons at entertainment and recreation facilities.

Typically, the RFID tags without batteries (i.e., passive RFID tags) are smaller, lighter and less expensive than those that are active RFID tags (i.e., powered RFID tags). The passive RFID tags are maintenance free and have lifetimes that are generally very long when compared to the requirements of their specific application. Additionally, considering all RFID tags, costs can range from less than 50 cents to more than $150.00 depending upon features and functionality. The powered backscatter RFID tags, also often referred to as semi-active RFID tags, represent a middle ground between passive RFID tags and active RFID tags. An example of such powered backscatter RFID tags is a PowerID Read Only 2000 tag by Power Paper Ltd. of Burr Ridge, Ill.

Even less complex RFID tags can be encoded with 64 or more bits of data that represent a large number of unique identification numbers. A 64-bit RFID tag can provide up to 18,446,744,073,709,551,616 unique identification numbers. An important attribute of some RFID systems is that a number of RFID tags can be interrogated simultaneously. This is a result of the signal processing associated with the technique of impressing the identification information on a carrier signal.

One attribute of RFID tags that is counter-intuitive is the ability of the unpowered, less complex RFID tags to support interrogation of multiple RFID tags at the same time. Using anti-collision techniques, the practical limit of simultaneous interrogations can be quite large. Using an anti-collision algorithm, multiple RFID tags can be readily identified, and even at the extreme reading range, the RFID tags require only minimal separation (e.g., five centimeters or less) to prevent mutual de-tuning. Most other identification systems such as bar code scanning require that only one device be interrogated at a time. The ability to deal with a plurality of devices simultaneously, and the ability to have them very closely spaced, and not physically viewable are desirable attributes for applications requiring rapid interrogation of a large number of items. In theory, it is one of the most useful attributes of RFID systems.

For example, a pallet of packaged consumer goods typically contains about 64 cubic feet of product cartons. Within this volume, it is not uncommon to have more than 100 cartons of product. A retailer may require that each carton, and the pallet which carries the stack of cartons, have an RFID tag affixed thereto. More than 100 other RFID tags, all of which are closer to the interrogator, may surround the innermost RFID tag in this instance. Thus, dealing with a plurality of devices simultaneously, and being able to have them very closely spaced are desirable attributes for applications requiring rapid interrogation of a large number of items. Unfortunately, conventional reader techniques will often make the reliability of interrogations of the innermost RFID tags described above less than acceptable.

One of the most powerful methods in the prior art is to combine one or more techniques for communicating with a plurality of RFID devices simultaneously. One such example is the use of anti-collision techniques and the “persistent sleep” function as described by the Electronic Product Code (“EPC”) and, more specifically, EPCglobal release EPC Specification for Class 1 Gen 2 RFID Specification, December 2004, and “Whitepaper: EPCglobal Class 1 Gen 2 RFID Specification,” published by Alien Technology Corporation, Morgan Hill, Calif. (2005), both of which are incorporated herein by reference. Anti-collision features allow an interrogator to increase the number of RFID tags it can observe. Having observed the RFID tags, the interrogator can command the RFID tags, which have been reliably read, to “sleep” so they will not respond to subsequent interrogations. The system then interrogates again, and is able to communicate with the RFID tags which were not observed in the first sequence. Though not truly simultaneous, the persistent sleep function of the EPC RFID tags permits a rapid and nearly simultaneous interrogation operation to be performed so long as the interrogator can reliably read the RFID tag.

In theory, it would seem that existing techniques would allow interrogation of an almost infinite number of RFID tags simultaneously, or nearly simultaneously. A number of practical issues, however, limit the theory of limitless numbers of simultaneous interrogations. For example, even with the frequency hopping commonly employed and using a conventional receiver, it is not possible to effectively filter the plurality of responses into individual signals or individual channels. As a result, an attempt to detect the response of any individual RFID tag in the presence of many other responses is often limited by interference. The effective signal-to-noise ratio at the interrogator is reduced when there is a plurality of responses, such that any individual response is less likely to be properly detected. The degree of signal-to-noise ratio degradation is dependent on a number of factors. While techniques such as persistent sleep can mitigate this problem, they do not sufficiently address the signal-to-noise problem to reliably receive information from the RFID tag in difficult environments. If a sufficient number of RFID tags are all closely grouped so that the effective signal-to-noise ratio is substantially reduced and, therefore, interfering with each other, reliable reading will be very difficult and, even if possible, will take a substantially longer amount of time than if these conditions were not present.

In the case of active or powered RFID tags, additional problems can arise. In particular, the longer reading ranges associated with these RFID tags can allow a very large number of RFID tags to respond at the same time, because the number of possible RFID tags is roughly proportional to the square (or even cube) of the interrogation range. Also, in the case of powered RFID tags, power management and avoidance of mutual interference has long been a topic familiar to those skilled in the art, as well as the subject of a number of Federal Communications Commission (“FCC”) actions.

The attributes of conventional systems, capable of interrogating RFID tags and also capable of interpreting the RFID tag responses are familiar to those skilled in the art. Each interrogator antenna is connected to radio frequency and digital circuits, which include a “reader” capable of detecting the presence of an RFID tag. An example of such an interrogator is a Series 2000 reader from Texas Instruments of Dallas, Tex., which provides the radio frequency and control functions necessary to communicate with 134.2 kilohertz, half-duplex, frequency-shift keying tags. The Series 2000 reader sends an energizing signal to the RFID tag. The RFID tag modulates the radio frequency energy impinging thereon, the backscatter of which is received and decoded by the reader. The Series 2000 reader then transmits a digital representation thereof via a standard serial interface (e.g., RS232 or RS422/485).

Taken together, an interrogator antenna, and reader and control circuitry form an interrogator, with a digital interface capable of transferring data to a digital computer, or to a communication network. The reader circuitry does not have to be close to the antennas, and can be separated by well over 100 meters if necessary. Thus, it is not necessary to operate the reader circuitry within the same physical volume as the items being interrogated. This provides the opportunity for placing antennas and, depending on the frequency, minimal support equipment (e.g., amplifiers and/or A/D converters) near items being interrogated, even if that environment is hostile to the operation of ordinary electronic equipment. The use of autocorrelation techniques, multiple port interrogators, separation of sensitive portions of the interrogation system from harsh environments and anti-collision techniques, can provide marked improvements over conventional RFID reader technology.

Some of the aforementioned interrogation techniques, however, whether used alone or used in some combination of techniques, have limitations. For example, multiple port receivers may introduce more signal interference than would otherwise be encountered by a single aperture system. This can occur when interrogation from a diversity of directions elicits more RFID tag responses than would have been generated by a single aperture. If only one viable signal path between the RFID tag and the interrogator exists, then these extra responses generally mean an effective increase in total noise/interference, without any improvement in signal power. At some point, the number of RFID tags responding can create a noise/interference environment that, without matched filtering such as autocorrelation, will cause a failure of the RFID tag\'s signal to be detected. The autocorrelation systems, however, generally need prior knowledge that heretofore has not been available in a timely fashion. Alternatively, algorithms that successively turn the RFID tags off (as in “persistent sleep”) can solve some portion of the collision problem, but also can substantially increase the time for reliable queries and responses. Timeliness of signal processing is important, since interrogators are usually installed at discreet locations such as loading doors or along conveyor systems. The opportunity to perform an interrogation and receive a fully compliant response may be a period of less than a second.

It is important to note that true instantaneous reading of an RFID tag is rarely a necessity. The conventional systems, however, typically use the real time paradigms of the bar code scanning techniques. In other words, in the same manner as a bar code that is only read and processed at the precise time that the bar code is being scanned, conventional RFID systems do not employ a memory of prior reads. This failure to learn from history is a serious drawback in conventional systems.

The RFID tags may be compatible with a number of harsh environments. One firm using RFID tags, Pierrel-Ospedali, an Italian pharmaceutical company, employs a manufacturing process that requires products to be sterilized for a period of time at over 120 degrees Celsius. Products enter an autoclave mounted on steel racks. Other examples of RFID tags withstanding harsh environments can be seen in the use of such devices for veterinary and animal husbandry purposes. The RFID tags are used to identify millions of livestock animals around the world. These systems track meat and dairy animals, valuable breeding stock and laboratory animals. The RFID tags may be hermetically sealed, and operate over the life of the animal. Body fluids, temperature, mechanical shock, normal electromagnetic interference and radiation such as x-rays should not affect the programmed code within the RFID tag. It is expected that the RFID tags will not only survive these environments; they will operate properly in such environments. Although the RFID tags may be capable of operating in this hostile environment, the inherent signal processing limitations associated with conventional receivers, common to RFID systems, is such that these hostile environments often degrade signal-to-noise and signal-to-interference ratios to such an extent that the interrogator cannot reliably detect or decode the response from the RFID tag.

Accordingly, what is needed in the art is an interrogation system and related method to identify and account for all types of items regardless of the environment or application that overcomes the deficiencies of the prior art. The interrogation system should be capable of counting hundreds or thousands of RFID tagged items quickly and reliably while in a confined space (e.g., casino chips in situ within a chip carrier). As another example, it should provide a system that can inventory the contents of a pallet or tote containing a large number of items, without unpacking. The interrogation system should also be capable of addressing a large number of powered backscatter RFID tags or active RFID tags without encountering severe signal-to-noise or signal-to-interference ratio problems. The signal processing of an interrogator should increase the overall effectiveness of a query and response cycle by noting that interesting and useful patterns can be inferred by sampling minimal data. Then, once the existence of the pattern is established, additional signal processing can be deployed to drive a sequence to a successful completion. As a result, the patterns described above can be employed to obtain useful information.

Additionally, what is needed in the art is an interrogation system that can derive information by a combination of current observables and prior observables that increase the automatic features of the interrogation system for material handling, and reduce the number of incidents involving manual unpacking of carriers such as pallets, crates, bins, and totes, while accounting for a plurality of items simultaneously. The interrogation system should support robust operation under conditions, and in environments that would otherwise be hostile. In accordance therewith, the interrogation system should accommodate a wide variety of signal noise and signal interference conditions, a wide variety of material handing procedures, and provide for a class of pseudorandom noise codes with characteristics desirable for both data processing and signal processing.